virtual_ips/conntrackd is slower than modular/virtual_ips/tasks.yaml suppose

[Dmitry Burmistrov]

One of deployments in virtual environment fails due to this timeout.
Modular starts here:
2016-07-13 18:32:00 +0000 Scope(Class[Osnailyfacter::Virtual_ips::Conntrackd]) (notice): MODULAR: virtual_ips/conntrackd.pp

And finishes here:
2016-07-13 18:33:31 +0000 Puppet (debug): Processing report from node-1.domain.local with processor Puppet::Reports::Store

And there were no errors during execution.

It is longer than 120 seconds. VMs are running on the server in a DC and have 4gb RAM each. So I don't think this nodes are "not strong enough" - I suppose to raise timeout. I suggest 200 seconds.
I think same timeout can be set for 'modular/virtual_ips/virtual_ips'.

Pacemaker wrappers/scripts sometimes work not fast enough (they are a bit slow).

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/342197

[OpenStack Infra (hudson-openstack)]

Change abandoned by Dmitry Burmistrov (<email address hidden>) on branch: master
Review: https://review.openstack.org/342197

[Dmitry Burmistrov (dmburmistrov)]

Error appeared only once on virtual environment. Timeout could be caused by networking or load issues.
In case of similar regular failures this bug can be reopened.